Proper parsing of "respondeo dicendum quod"

Question

Anyone who has read the Scholastics, such as St. Thomas Aquinas, is familiar with the basic structure of an articulus: (Here's an example.)

1. Objections ("videtur quod...")
2. Quote from authority ("sed contra est...")
3. Corpus ("respondeo dicendum quod...")
4. Response to objections ("ad primum ergo dicendum quod....")

I am curious about the proper parsing of "respondeo dicendum quod." Although it's not surprising to see quod used to begin indirect discourse, I'm especially curious about dicendum. What exactly is it? Here are some possibilities:

1. It is an accusative gerund, the object of respondeo.
2. It is a neuter singular nominative impersonal gerundive, i.e. dicendum est quod

Neither makes a whole ton of sense to my mind, although it's pretty obvious what a translation should look like. In fact, many translations seem to translate as if it were an ablative, e.g. "I respond by saying that..."

Answer 1

For dicendum, it would be necessary to supply the ellipsis est, yielding dicendum est, a passive periphrastic construction. Altogether, the phrase respondeo dicendum quod would thus be translated into English as, “I respond: it must be said that...” Therefore, I would say dicendum is parsed as a neuter gender, singular number, nominative case, gerundive.

In other works, Aquinas uses the entire phrase «Respondeo dicendum est quod» quite frequently.¹

For example:

---

Footnotes

¹ I searched for the phrase «respondeo dicendum est quod» in the Corpus Thomisticum search tool, and it found “159 cases in 108 places.”

Answer 2

First of all, the use of the accusative gerund without preposition isn't verifiable with certainty until medieval Latin (LG II § 203 p. 377), despite AALTO pp. 82-85 quotes some passages of archaic texts which could attest this use (for example Pl. Asin. 338: iam devorandum censes; Ter Phor. 457: ego amplius deliberandum censeo; CIL I 2 1759 pontem peilas faciendum coiravere).

On the contrary, suppose a neuter singular nominative impersonal gerundive should be excluded as it doesn't suit for the case we are examining.

We then should take into account the possibility of a predicative gerundive, a construction attested for a large number of verbs (such as do, adduco, peto, rogo, concedo, defero, denoto, mitto, relinquo, suscipio, trado and, most of all, habeo) in classical authors, although used by medieval authors only (mainly Alexander of Hales, Bonaventura and St. Thomas) with the verbs respondeo and dico (vd. LG II § 202 [p. 372]).

Examples of this construction are:

• Ter. Phorm. 365: agrum ... colendum habebat;

• Varro rust. 1,13,2: multi ... habent ... importandum;

• Sen. contr. 2,7,1: confitendum habeo ... me ... fuisse.

We could add that in Italian the phrase "rispondo dicendo che" is a sort of syntactical unit used in prose as in every day language, and in that case the verb dicendo is predicative of rispondo.

So I would consider the verb dicendum a predicative gerundive of respondeo rather than an accusative gerund object of respondeo or a neuter singular nominative impersonal gerundive.

Bibliography:

• M. Leumann - J.B. Hofmann A. Szantyr, Lateinische Grammatik, München 1972-1979 (LG).

• P. Aalto, Untersuchungen über das lateinische Gerundium und Gerundivum, Helsinki 1949 (AALTO).